This stuff can work fine for small scale projects, one offs, etc. But placing this in the context of an alternative or replacement for major federal funding is deeply flawed.
1) Overhead rate. Now admittedly, not all Universities bother going after their indirect costs for small philanthropic donations. But if a lab tries to exist on this strategy? You can be damn sure they are going to come after indirects. Some Universities do this already. And donors don't like it. You can bet there's some fancy tapdancing trying to figure out how to minimize revealing to the medium ticket donors that their donation are getting taxed. The big ones fight it, obvs.
2) Chump change. Sorry but it is. Perlstein raised $25K. The NIH R03 is $50K for two years. The R21 is $275K over two years and the R01, as we've discussed, is most typically $250K (in direct costs, mind you) for 4-5 years. There is going to be very, very little that can be accomplished with the kind of cash that is available via crowdfunding.
3) Yeahbut! The uBiome and American Gut projects raised over $600K, man! Yeah, the former is at $286,548 and the latter is at $339,541 as of this writing. Impressive. Right? but the total is less than the cost of two years of NIH R01 funding. And these may be the best examples. Time will show how many of these can go viral and make big bucks, how many can get $25,000 and how many struggle to get $5,000. Color me extremely skeptical on the big-bucks ones.
4) Can it repeat? That's another critical question. All well and good for Perlstein to pull down $25K in crowdfunding. But he needs to do it again. and again. and again. No offense but crowd funding works the first time on novelty, your buddies and people looking to make a point. Think they'd be lining up to throw down for Perlstein's second project in such numbers? Will people who don't even know him flog the shit out of the Twitt stream like they did for his Meth study? Here's a hint: hell no.
5) Deliverables. Part of the problem is the nature of the deliverables. What is the crowd to see that has been done with their money? Well, from Perlstein's project description, the data are going up online as they roll in. So...figures. basically. Not even clear that there will be a pub on which they can be acknowledged. The small scope of the project make it likely that at best one publishable panel will result. And dude, will regular journals put up with the Supplementary Acknowledgement Table approach so all donors can be listed? maybe. but what, now you are going to return to the same crowd and say "Hey, throw down another $25K and we'll do Figure 2...if I still have a job, that is".
6) Science is a tough sell. Still. It is extremely difficult to see where anything Perlstein happens to find about the intracellular distribution of methamphetamine is going to so engage the crowd that it jumps in with more funding. This is pretty basic science. It would take "I am mere inches away from curing Meth addiction" level stuff to grab the crowd if you ask me (and anyway, if you did that, Pharma would come a'callin'). In contrast, I offer the outcome for one of my favorite scifi authors. Tobias Buckell had a decent fan base, a book series for which there was a clamor for more from his crowd and he was asking for a mere $10K. He raised it, wrote the book and delivered that sucker to his readers (Kickstarter backers and nonbackers alike). It was, to my read, the same book he would have written (and I would have purchased) if he'd had a schweet advance deal at a major publisher. Or if he'd (somehow) still been able to write on spec like a noob author. Same damn product. Can we say the same for a $25k SCIENCE project? I think not.